Detachment of compliant films adhered to stiff substrates via van der Waals interactions: role of frictional sliding during peeling

Abstract

The remarkable ability of some plants and animals to cling strongly to substrates despite relatively weak interfacial bonds has important implications for the development of synthetic adhesives. Here, we examine the origins of large detachment forces using a thin elastomer tape adhered to a glass slide via van der Waals interactions, which serves as a model system for geckos, mussels and ivy. The forces required for peeling of the tape are shown to be a strong function of the angle of peeling, which is a consequence of frictional sliding at the edge of attachment that serves to dissipate energy that would otherwise drive detachment. Experiments and theory demonstrate that proper accounting for frictional sliding leads to an inferred work of adhesion of only approximately 0.5 J m(-2) (defined for purely normal separations) for all load orientations. This starkly contrasts with the interface energies inferred using conventional interface fracture models that assume pure sticking behaviour, which are considerably larger and shown to depend not only on the mode-mixity, but also on the magnitude of the mode-I stress intensity factor. The implications for developing frameworks to predict detachment forces in the presence of interface sliding are briefly discussed.

Keywords: digital image correlation; friction; peel test; poly (dimethylsiloxane); slip; thin-film adhesion.

Figure 1.

(a) Schematic of the peel test set-up. A speckled PDMS film is adhered to a glass slide with one end clamped to a load cell fixture. The load point is moved via a linear stage at a constant angle (ϕ); the film angle (θ) changes as the test progresses. (b) Displacements in the adhered region of the film near the crack front are measured via DIC. (Online version in colour.)

Figure 2.

Peel force versus peel angle for (a) 50, (b) 80 and (c) 125 μm thick films. At angles approximately less than 30°, the data are in much better agreement with predictions for the critical peeling force from the present ‘sliding’ model (solid and dashed black lines for the small and large deformation models, respectively) as compared to Kendall's ‘sticking’ model coupled with a constant adhesion energy. In all cases, experimental error bars are smaller than the plotting points. The toughness values shown correspond to least-squares fits of the data to the small deformation sliding model, for each film thickness. (Online version in colour.)

Figure 3.

Energy release rate versus peel angle predicted from sticking and sliding models of peeling for (a,d) 50, (b,e) 80 and (c,f) 125 μm thick films. In (a–c), the energy release rate values predicted by sticking and sliding models diverge at intermediate angles (approx. 20°). (d–f) On a smaller scale, the energy release rates predicted from large deformation (open symbols) and small deformation (filled symbols) sliding models, highlighting the similarity in calculated values at all but the smallest peel angles. (Online version in colour.)

Figure 4.

Lateral surface displacements determined via DIC for (a) 50, (b) 80 and (c) 125 μm thick films, measured at peel angles of approximately 9°. The solid line represents a least-squares fit of the data to an exponential decay function. For similar peel angles, characteristic decay length increases with increasing film thickness. (d) Image of a representative region near the crack front used to determine the displacement profile for the 50 μm thick film data in (a). (e) Contour plot of lateral displacement corresponding to the image in (d). The dashed line shows the approximate position of the crack front. White regions represent areas in which image correlation failed, leading to measurement uncertainty in the displacement near the edge of delamination.

Figure 5.

Measured displacements in the adhered region of the film near the crack front as a function of peel angle and film thickness. The dashed lines correspond to the predicted decay distance for a fully-sticking film, for each film thickness.

Figure 6.

Mixed-mode fracture analysis of peeling. (a) Mode-I and mode-II stress intensity factors calculated from equations (4.1) with experimental peel forces, as a function of peel angle. (b) Phase angle as a function of peel angle and (c) apparent interface toughness as a function of phase angle. The symbols represent values calculated in a domain allowing −180° < ψ <−90°, whereas the dashed lines correspond to the convention of defining ψ =−90° when K_I < 0.

Figure 7.

The uniaxial response of a speckled PDMS film is well characterized by a neo-Hookean constitutive law. (Online version in colour.)